White matter dysconnectivity is common to both schizophrenia and psychotic bipolar disorder, independent of current symptoms and shared by their first-degree relatives. Measures of brain connectivity may therefore potentially serve as an intermediate phenotype for psychotic disorders that more closely reflect the biological effects of illness susceptibility compared to the clinical phenotype. The study of unaffected siblings is a powerful design to assess genetic risk for psychosis with evidence of substantial heritability in moderately large family studies of healthy individuals. Tractography using diffusion tensor imaging (DTI) has revealed abnormalities in fractional anisotropy (FA), especially within the superior longitudinal fasciculus (SLF) both in schizophrenia and psychotic bipolar disorder. In addition, own pilot data in 224 individuals using probabilistic DTI tractography provide evidence for lower FA within the SLF both in first-episode and chronic psychosis patients compared to age- and sex-matched healthy volunteers with similar effect sizes across patient groups suggesting that these abnormalities may be a trait-related manifestation of psychosis. Moreover, abnormalities in this tract are evident among individuals at risk for developing psychotic disorders and strongly predict deficits in neuropsychological and social/role functioning making it a potentially important target for investigation and intervention. A limitation of diffusion tensor based tractography, however, is that it can be unreliable and lead to misleading information in regions of crossing fibers. We therefore believe the field needs to move beyond the diffusion tensor framework to ultimately provide clinically and biologically relevant information to the treating clinician. In contrast to diffusion tensor imaging, diffusion spectrum imaging (DSI) enables resolution of crossing fibers, but requires significant scanning time that is not typically feasible in clinical populations. In the current study, we propose using compressed sensing to accelerate DSI to investigate crossing white matter fibers and additional DSI parameters in regions of crossing white matter fibers in 25 antipsychotic drug-nave first-episode psychosis patients, 25 of their unaffected siblings and 25 healthy volunteers. The overarching goal of the study is to examine whether novel indices of putative white matter integrity using DSI can serve as potential endophenotypes for psychosis independent of diagnostic category to inform the molecular basis of psychotic phenomenology without the confound of antipsychotic medication. Moreover, identification of such measures in unaffected siblings of patients with psychosis could serve as biomarkers to gauge the effects of novel interventions in future studies.